Re: [PATCH] statfs64 no longer missing

[Peter Chubb]

>>>>> "Andi" == Andi Kleen <[e-mail]> writes:

>> --- a/arch/i386/kernel/entry.S Fri Oct 18 10:07:29 2002 +++
>> b/arch/i386/kernel/entry.S Fri Oct 18 10:07:29 2002 @@ -737,6
>> +737,8 @@ .long sys_free_hugepages .long sys_exit_group .long
>> sys_lookup_dcookie + .long sys_statfs64 + .long sys_fstatfs64 /*
>> 255 */

Andi> Funny. Finally filling the 8bits for syscall numbers.

Yes. 

>> +long vfs_statfs64(struct super_block *sb, struct statfs64 *buf) +{
>> + struct kstatfs st; + int retval; + + retval = vfs_statfs(sb,
>> &st); + if (retval) + return retval; + + if (sizeof(*buf) ==
>> sizeof(st)) + memcpy(buf, &st, sizeof(st));

Andi> Don't you need to clear spare here too ?

I don't think so -- it's vfs_statfs() where *buf is zeroed --- the
memcpy copies the zeroed spare[] fields too.

Peter C